CSE 5346 Networks II High Performance Networks 1

CSE 5346 – Networks II: High Performance Networks 1

Introduction Performance and Quality of Service 2

Course Foundation Assumptions (pre-requisites) l l l Good understanding of packet-switched networking concepts and principles of operation Good understanding of Internet protocols and architectures (e. g. , IP protocol stack) Solid foundation in computer operating systems fundamentals Ability to learn simple programming languages (such as ns/2 tcl, C/C++) Ability to perform independent research, analyze findings and document results Introduction 3

What will we cover? Some Review l l New Content l l l Networking Introduction/Review – Today… setting the stage Frame Relay, ATM & High Speed LANs Performance Modeling and Estimation Congestion, Network Traffic Analysis/Management and Routing Quality of Service Compression Standards and Practice Advanced Topics (time permitting): Vo. IP, Mobile Agents Introduction 4

How will we cover it? Classroom lectures l Quizzes and exams l Incremental modeling and simulation project using ns/2 l Self-study research, and simulation reports by students l Study and student presentations on relevant papers & RFCs l Introduction 5

Chapter 1 - Introduction An Overview of Networking the Need for Speed and Quality of Service 6

Growth of the Internet As of January 2005 … > 300 million “computers” in 209 countries Introduction 8

What is an/the Internet? l connected computing devices: hosts, endsystems – PC’s, workstations, servers – PDA’s, phones, toasters, cars l l running network applications communication links – fiber, copper, radio, satellite routers/switches: forward packets (chunks) of data thru network Introduction router server local ISP workstation mobile regional ISP company network 9

The Need for Speed! l Scale – growing number of hosts -> growing demands on bandwidth – new technologies result in new paradigms for device and connection types l l e. g. ? ? User Expectations! Introduction l Application – demand for large to huge file transfers – increasing critical nature of Internet use – demand for “realtime” performance characteristics – demand for guarantees of service levels l e. g. ? ? 10

High-Speed Networks: IDN to ATM l IDN (Integrated Digital Network) l ISDN (Integrated Services Digital Network) – early ’ 60 s, answer to growth of digital, computer-controlled, circuit-switched networking – WE 4 ESS introduced in 1976, 1 st large scale commercial time-division switch – integrated voice and data on the same digital transmission links/exchanges – small incremental cost for data over existing digital voice network Introduction 11

High-Speed Networks: IDN to ATM l Frame Relay l B-ISDN (Broadband ISDN) – popularized standard (c. 1988) for packet switching over ISDN – most widely deployed WAN technology in use today – c. 1988: emerging demand for broadband services – new high-speed technologies available – emerging bandwidth hungry applications Introduction 12

High-Speed Networks: IDN to ATM l ATM (Asynchronous Transfer Mode) – early ’ 90 s outgrowth of emerging need for highspeed switching over B-ISDN WAN – rapidly evolved as high-speed packet switching technology of its own accord – primary deployment today is: l l l public network infrastructure LAN backbone private network, VPN WAN – services not widely deployed/available until late -’ 90 s. Introduction 13

High-Speed Networks: ATM Backbone Example Introduction 14

High-Speed Networks: LANs l High-speed LANs – driven by explosive growth in speed and computing power of PCs in 1990 s – emergence of client-server computing architecture – use of centralized server farms – emergence of “power workgroups” and workgroup applications – need for local high-speed LAN backbones Introduction 15

IP-based Internets (aka TCP/IP networks) Internetworking: the dominant paradigm of computer networking Evolution - key internetworking technologies: l l packet switching TCP/IP – TCP: reliable end-to-end transport – IP: internet routing and delivery dynamic routing, load balancing high speed Ethernet LANs Introduction 16

The Need for Improved (better) Levels of Service l Internet Best. Effort Service l Applications – often create inelastic traffic – often sensitive to delay – often sensitive to jitter – often critical in nature – generate elastic traffic as well – all packets treated equally – designed for elastic traffic – no guarantees of bandwidth or throughput – no guarantees of delay – no guarantee of jitter (delay l User Requirements! variation) Introduction 18

Delay Sensitivity & Criticality Introduction 19

Delays in Packet Switched (e. g. IP) Networks l End-to-end delay (simplified) = (dprop + dtrans + dqueue + dproc) x Q A B Where: Ø Propagation delay (dprop) Ø Transmission delay (dtrans) Ø Queuing delay (dqueue) Ø Processing delay (dproc) Ø Number of links (Q) Introduction 20

So what’s the problem? What makes this so hard? (I. e. , what are we going to focus on in this course. ) Introduction 21

Delays in Packet Switched (e. g. IP) Networks l A l End-to-end delay (simplified) = – (dprop + dtrans + dqueue + dproc) … on each link Where: B Ø Propagation delay (dprop) = d/s Ø Transmission delay (dtrans) = L/R Ø Queuing delay (dqueue) = ? Ø Processing delay (dproc) = ? More later Ø Number of links (Q) = ? Introduction 22